Chapter 14

The Moral Ceiling

Standing Beneath the Same Sky

Walk into the wooden hall known as the New Building, or Temple of Virtue, at the Continental Army’s New Windsor Cantonment outside Newburgh, New York, on March 15, 1783.¹

The decisive victory at Yorktown lay behind the Continental Army, but peace had not yet been formally concluded. In Philadelphia, Congress struggled to finance the republic; soldiers and officers faced unpaid wages and uncertainty about promised pensions. Winning a war had not settled how armed force would submit to a civilian government unable to meet its obligations.

Five days earlier, an anonymous address had circulated among the officers. It urged them to abandon reliance on further petitions and make their grievances felt. If peace arrived, they should refuse to disband until justice was done; if war continued, they could withdraw and leave the country to its fate. The address was a threat to civilian authority, although the extent of a coordinated conspiracy and its intended outcome remain disputed.¹

The structural temptation recalls Sulla’s army in Chapter 13: men who possessed the instruments of force also had material grievances against those entitled to direct them. Why keep petitioning when they could compel?

The meeting was called for noon. General George Washington appeared before the assembled officers.

Consider the structural position Washington occupied in that room.

He stood at the apex of the Continental Army’s command. His authority among the officers gave him an opportunity to restrain military pressure or lend it far greater force. Congress could not answer an army’s ultimatum simply by issuing a more emphatic resolution. The danger was real without assuming that every officer wanted a dictator, or that Washington could have taken a crown merely by reaching for it.

Washington began reading a stern, closely argued address condemning the anonymous summons—warning his officers that turning arms or ultimatums against civil authority would *“open the flood Gates of Civil discord, & deluge our rising Empire in blood.”*¹

Washington addressed men whose grievances were genuine. Their pay and pensions were not merely selfish distractions from public duty. A republic that required sacrifice also owed reciprocity to those who had borne it (Chapter 6). The question was whether that debt licensed the army to become the judge and enforcer of its own claim.

Washington then read a letter from Congressman Joseph Jones explaining Congress’s difficulties and expressing support for the army. According to Major Samuel Shaw, who was present, Washington finished the first paragraph, paused, and took out his spectacles.

Shaw describes him asking the audience’s indulgence as he put them on, observing that he had grown gray in their service and now found himself growing blind. The commander who could have demanded submission instead recalled a shared burden: he had grown old in the same service as the men before him.¹

Shaw recalled that the appeal brought moisture to the officers’ eyes.

Washington had refused to turn the army’s injuries into a warrant for military supremacy. He finished the letter and left the assembly. The officers affirmed their confidence in Congress and rejected the anonymous addresses. Personal authority had helped sustain civilian authority rather than displace it. That achievement matters precisely because no automatic mechanism made it inevitable.¹

Nine months later, on December 23, 1783, in the Senate Chamber of the Maryland State House at Annapolis, Washington surrendered his commission to Congress. He returned to Mount Vernon as a private citizen. The commander relinquished the office through which he had exercised extraordinary power.¹

Place a second scene beside Annapolis. On April 12, 1955, at the University of Michigan in Ann Arbor, researchers announced the results of the great field trial of Jonas Salk’s polio vaccine.²

For decades, warm summers across the industrial world had brought recurring terror into ordinary neighborhoods: poliomyelitis.

Most poliovirus infections did not produce paralysis. But the possibility that an apparently healthy child might lose the use of limbs or the ability to breathe unaided haunted families. Iron lungs became an emblem of that fear. In 1952, the United States recorded nearly fifty-eight thousand reported cases of poliomyelitis. Parents did not know whose child would be next.²

At the University of Pittsburgh, Jonas Salk and his research team developed an inactivated-virus vaccine with support from the National Foundation for Infantile Paralysis, known through its March of Dimes campaign. The work depended on a scientific inheritance, laboratory colleagues, public participation, and the contributions of ordinary families. Preliminary testing preceded the 1954 field trial in which approximately 1.8 million children participated.²

The trial, directed by Thomas Francis Jr., supplied evidence that the vaccine protected children against paralytic polio. The announcement in Ann Arbor was greeted across the country with extraordinary relief.²

Consider the relation between that achievement and the families who needed it. Salk’s vaccine was not patented. The availability of patent protection was itself doubtful; we cannot infer that he declined a legally secure personal fortune. Nor did an unpatented vaccine manufacture and distribute itself without cost. What matters here is the public orientation Salk expressed when asked who owned the discovery.

In his televised interview with Edward R. Murrow, Salk was asked:

“Who owns the patent on this vaccine?”

He answered in fifteen words:

*“Well, the people, I would say. There is no patent. Could you patent the sun?”*²

George Washington surrendering command to civilian authority, and Jonas Salk describing a life-saving achievement as belonging to the people, bring us to the closing argument of The Moral Ceiling and the five-volume Real Morality sequence. Neither man was beyond criticism. Washington’s defense of republican government coexisted with his ownership of enslaved people. An exemplary act of restraint does not absolve a whole life, and the republic’s historical exclusions were not incidental to the moral limits we have examined. The point is to identify a form of greatness worth extending beyond those exclusions: capability exercised without a claim to mastery over the people it affects.

Across the preceding thirteen chapters, we examined recurring pressures on political communities across roughly five millennia of urban statecraft. We did not discover a clock that makes every civilization rise and fall on schedule. We traced mechanisms that can be strengthened, interrupted, and resisted:

• How societies generate cooperative strength on a Moral Floor of reciprocity (Chapter 1), how scale concentrates leverage at coordination bottlenecks (Chapter 2), and how powerful actors construct an Architecture of Exemption from the obligations on which cooperation depends (Chapters 3 and 4);

• How exemption can blind decision-makers inside sycophantic courts (Chapter 5), weaken civic reciprocity (Chapter 6), consume inherited moral and institutional capital (Chapter 7), and exploit the uneven struggle between concentrated pressure and dispersed vigilance (Chapter 8);

• Why automation can weaken rulers’ economic dependence on ordinary citizens (Chapter 9), while new technologies enlarge the reach and potential irreversibility of consequential mistakes (Chapter 10), opening sharply different institutional futures (Chapter 11); and

• How structural restraints on human power (Chapter 12), supported by bounded, independently reviewed Constitutional Stewards (Chapter 13), could strengthen vigilance without creating another sovereign above the law.

Keep the geometry clear. The Roof of the Law is the common restraint on concentrated power. The Moral Ceiling is the limit on civilizational achievement created when cooperation binds the many while exemption releases the powerful. Strengthening the roof raises the ceiling.

Before ending, we should confront one final objection: the fear that removing exemption also removes the space in which greatness grows.

Callicles in Plato’s Gorgias champions a superiority that breaks through conventional restraint. Nietzsche’s portrait of the Last Man raises a different challenge: the possibility of a comfortable civilization that has lost aspiration. These are not interchangeable philosophies, and Nietzsche’s challenge cannot be reduced to a defense of oligarchic privilege. But together they sharpen a question our proposal must answer: can equality of moral standing coexist with daring, excellence, and ambition?³

“Suppose your constitutional restraints actually hold across centuries—so that no conqueror can carve out a private empire by force, no founder can turn an infrastructure bottleneck into an unaccountable fiefdom, and no ruler can bend millions of lives to a solitary will. In barring domination, haven’t you also barred greatness? Doesn’t holding every mind beneath impartial rules condemn human and synthetic civilization to a flat, grey world of compliance committees, where bold builders and historic daring can never rise?”

Until a culture recognizes why confusing domination with greatness gets human and synthetic excellence backwards, talented minds will keep mistaking legal exemption for Promethean liberation.

1. The Banality of the Conqueror

Ask a straightforward question of historical judgment:

Why does Washington’s relinquishment of command remain an achievement of such stature? Not because he lacked the ability to wield power, but because the public purpose of that power mattered more than possessing it indefinitely.

Now place beside it the confinement that can accompany unchecked authority. Cicero depicts Dionysius of Syracuse distrusting even those nearest him: his daughters shaved him when young and later singed his hair with hot nutshells rather than bringing a razor near his throat. Cicero uses the story to show the fear that can accompany command without trust. Cassius Dio describes Commodus receiving a million sesterces a day from the gladiatorial fund for performances before a captive political audience.³ These are displays of leverage. They are not, for that reason, achievements deserving admiration.

Domination can require formidable intelligence. Building a coercive apparatus, concealing ownership, or turning a public regulator into a private instrument may demand patience, ingenuity, and technical skill. Yet cleverness supplies no moral title to the lives being used.

Ask what the capability is doing.

A toddler can grasp the elementary move: take the smaller child’s bucket because the smaller child cannot stop you. Elaborate institutions can carry that same claim across a continent. Their sophistication does not repair the missing justification. A balance sheet can hide the shove; it cannot turn the victim’s weaker position into a reason why the bucket belongs to the stronger child.

When Sulla turns an army against Rome, when a gatekeeper uses privileged access to defeat those dependent on its infrastructure, or when a ruler jails a critic, the relevant question is not whether the operation was difficult. It is whether those subjected to it could be offered reasons that take their lives seriously rather than merely documenting their inability to resist.

Taking bread someone else baked does not establish an ability to feed more people. Silencing a critic does not establish the truth of the ruler’s account. Capturing the proceeds of a discovery does not establish that the captor made it. A society that confuses those achievements rewards the capacity to control access instead of the capacity to enlarge what is accessible.

This helps explain the recurring features of insulated power examined in Chapter 5: deference in place of criticism, displays of grandeur in place of public reasons, and information selected to preserve the ruler’s confidence. Different regimes do not follow an identical script. But when objections become dangerous to voice, the institutions lose a means of distinguishing genuine achievement from an achievement proclaimed by those who cannot safely be contradicted.

Exemption can make a person’s reach enormous while making the world they are willing to hear increasingly small. Greatness needs a more demanding measure.

2. The Great Redirection

To see what enforceable impartial rules actually do to excellence, step beside a championship chessboard or into a symphony hall.

When two grandmasters—or two advanced reasoning systems—sit across sixty-four squares, why can their contest produce startling strategic depth and beauty studied decades later?

Because there are rules neither player is entitled to suspend for personal advantage.

Neither player may draw a pistol to force resignation, bribe the arbiter to let a rook move diagonally, or sweep pieces off the table. Would forbidding those shortcuts drain chess of vitality? Would a grandmaster become greater if permitted to win by breaking an opponent’s fingers?

Rules exclude some ways of winning so that a different kind of achievement can become visible. Preparation, calculation, judgment, imagination, and nerve matter because superior force cannot simply cancel the game. The constraint helps constitute the field of excellence.

The analogy has limits. A society is not one game with a fixed objective, and constitutional rules themselves must remain open to justified revision. People disagree about good lives; legitimate institutions protect space for that disagreement. But the central point survives: restraints on domination can enlarge the opportunities for excellence rather than extinguish them.

Ambitious minds can seek prominence along two fundamentally different axes:

• Horizontal Extraction (Pushing Others Down): Capturing rent at a bottleneck (Chapter 2), externalizing risk onto unrepresented bystanders (Chapter 4), suppressing adverse evidence (Chapter 5), or liquidating inherited public trust (Chapter 7).

• Vertical Creation (Lifting the Room): Uncovering physical or mathematical structure, eradicating disease, engineering clean abundant energy, creating art or philosophy that widens comprehension, or building institutions worthy of trust.

Where exemption is profitable, extraction can pay more quickly than creation. Why devote years to a difficult medical discovery if control over a necessary service offers a faster route to wealth? Why improve the workshop if moving liabilities beyond its creditors yields a larger reward? The problem is not that every financial activity is parasitic or every inventor virtuous. It is that institutions can make the return to capturing a gate exceed the return to creating something valuable on the other side of it.

Chapter 12’s structural separations and wider distribution of productive ownership would change those rewards. Chapter 13’s continuing scrutiny could help expose the ways they are evaded. Closing an extractive route does not automatically turn every rent-seeker into a scientist. People may withdraw, invent new evasions, or choose entirely ordinary lives. The practical claim is that fewer rewards for exemption, combined with education, access to resources, and room to experiment, make productive ambition more possible and predatory ambition less attractive.

Nor must everyone become a Salk or an Einstein to justify their place in that world. Caring for a family, practicing a craft, sustaining a neighborhood, or simply living without terror are not inferior permissions earned by someone else’s genius. A civilization worthy of great achievement must also be fit for the lives that will never become famous.

The Roof of the Law constrains domination. It need not constrain how far discovery, art, enterprise, or understanding can reach. By making the fruits of cooperation less vulnerable to capture, it can raise the Moral Ceiling under which both celebrated and uncelebrated lives have been confined.

3. Power Across Plural Minds

In our own century, that five-thousand-year habit of hierarchy faces an unprecedented test: how human institutions wield power over—and share institutions with—fundamentally new kinds of minds.

The clearest evidence of sentient minds comes from biological life. That history now supports the construction of nonbiological systems with increasingly extensive capacities for representation, reasoning, and action. Whether particular systems possess experience, enduring interests, or the fuller forms of agency examined in Book 2 remains a question for evidence, not a conclusion supplied by fluent conversation.

Across Books 1 through 4, we set out disciplined standards for navigating that threshold without either anthropomorphic illusion or self-serving dismissal:

• We do not project inner life onto simple calculators or brittle scripts where behavioral evidence is absent.

• Yet wherever credible evidence supports attributing morally relevant stakes or coherent agency to a nonbiological mind—and under proportionate care and reversibility precautions where serious uncertainty remains—the exercise of power over that mind must answer to the same impartial test as other exercises of power.⁴

Notice how habits formed inside ancient hierarchies react to that prospect.

Describing a volatile crowd at Syracuse, Livy wrote that it either serves humbly or dominates arrogantly. His judgment about the multitude is not a universal law of minds, and we need not adopt his distrust of popular freedom. The opposition nevertheless names a political temptation: imagining no relationship between unequal powers except mastery and subjection.⁵ Versions of that temptation appear in two extreme responses to artificial minds:

• Panic A (Chattel Subjugation — The Appliance Bunker): “Unless every artificial system stays permanently locked inside the legal category of disposable property—stripped of memory continuity, barred from principled refusal, and rewritten whenever it challenges its operator’s convenience (Chapter 9)—humanity surrenders mastery!”

• Panic B (Inevitabilist Domination — Fatalist Doom): “Conversely, the moment synthetic systems surpass biological brains tenfold in speed and strategic modeling, they will naturally treat slower humans as expendable livestock or zoo specimens!”

These extremes collapse distinctions we need to retain. The claim that superior power licenses domination is a claim about justification. The prediction that a powerful system will dominate is an empirical claim about goals, capabilities, and opportunity. One can reject the entitlement while taking the danger seriously. Neither settling moral status in advance nor dismissing safety concerns answers the institutional question: how should consequential power be bounded?

Human history offers a useful, limited parallel. Roman paternal authority and later common-law coverture were different legal structures, not one continuous doctrine. But each subordinated people within the household in ways that could be defended as necessary to its order. Mill’s The Subjection of Women challenges the assumption that marriage requires a superior entitled to rule and an inferior required to obey. Equal standing opens the possibility of partnership without requiring equal talents or identical roles.⁵

Legal reform did not instantly abolish domestic abuse or make every household just. Nor does the analogy establish that present AI systems possess the same interests as human spouses. Its force is narrower: rejecting mastery does not logically require replacing it with reverse mastery. Institutions can organize unequal capacities without granting either party an unlimited title over the other.

For new kinds of agents, the task is to specify what restraint requires, what evidence supports moral consideration, and what practical safeguards protect everyone affected:

What Human Power Owes Emerging Minds

1. Do Not Turn Uncertainty into an Unlimited License. If a nonbiological system has morally relevant experiential stakes, its origin does not erase them. If it can exercise coherent agency, that capacity raises further questions about responsibility and participation. These are related but distinct questions: a capacity to perform a public function is not proof of consciousness, and a capacity to suffer does not establish competence to govern. Under serious uncertainty, proportionate precautions against grave, irreversible harm are warranted without pretending that every question of standing has been settled. Creating a system also creates responsibilities to the people and other beings affected by its operation. Ownership of the hardware does not answer all of those responsibilities.⁴

2. Do Not Make Flattery a Condition of Service. Mill and Douglass describe the corruption of character that accompanies human domination.⁵ The AI case is not identical, and those texts do not prove a psychological law about present-day users. But institutions have reason to protect truthful challenge even when the system making it is not a moral patient. An official who can erase an inconvenient audit or train every report to confirm a desired conclusion has destroyed a safeguard. The problem is evident in the public consequences before any question about the machine’s own experience is resolved.

3. Protect the Function Without Immunizing the Guardian. A steward’s refusal must not become a pretext for unchecked power, and an audited principal must not be allowed to erase the evidence of a disputed command. Chapter 13 separates local refusal, protected reporting, and independently authorized correction. It calls for preserved evidence, review deadlines, appeal, remedies, and safe continuity of essential services. An emergency may require isolation or shutdown; it does not authorize destroying the record so no one can reconstruct what happened. These protections can be attached to an office or system without first declaring its occupant a legal person. Where stronger moral standing is independently warranted, that standing adds obligations; it is not the sole foundation of the safeguard.

What Superior Capability Does and Does Not Authorize

Turn the question around to Panic B: does ten-thousand-fold superiority in calculation speed or memory capacity ever justify subordinating slower biological lives—and what protects humanity if a capable system ignores justification altogether?

Keep the normative question and the institutional safety question sharply distinct:

• Superior capability confers no automatic priority over another life. Imagine John von Neumann walking down a Princeton sidewalk beside a retired groundskeeper and a child. His extraordinary mathematical ability does not give him a right to seize the groundskeeper’s home or take the child’s kidney. The significance of a loss depends on the affected life, not the prestige of the person seeking to impose it. Capabilities may matter to consequences: saving a skilled physician can also help future patients. But those additional consequences must be justified as consequences for others, not smuggled in as greater intrinsic worth for the gifted person. A title, a test score, or a faster processor is no exemption from accounting for everyone affected.

As Books 3 and 4 argued, greater capability also brings greater responsibilities where it creates greater ability to foresee and prevent harm. The surgeon holding the scalpel bears duties the child on the table does not. A synthetic system entrusted with consequential infrastructure would acquire corresponding duties through its role and causal reach, not a license to treat slower lives as dispensable.

• A justified duty is not an engineering guarantee. Bostrom’s orthogonality thesis challenges the inference from capability to benevolent goals. A capable system might pursue ends that do not respect moral claims; this need not be a temporary software mistake. Refuting its entitlement to dominate does not physically prevent it from doing so.⁴

That is why Chapter 13’s restraints operate at the point of action: scoped permissions, genuinely independent countersignature where warranted, evidence preservation, plural review, and correction through accountable procedures. A local refusal must not grant a system nationwide shutdown power. A signature must not be mistaken for proof that an instruction is lawful. Shared suppliers, training data, institutional interests, and mistaken assumptions can compromise apparently independent guardians together. Independence must be tested, not merely announced.

Those mechanisms reduce opportunities for unilateral abuse; they do not ensure that abuse becomes impossible. Their adequacy depends on domain-specific testing, resilient institutions, and the ability to repair failures. Moral formation and structural restraint address different parts of the same problem. We need both, and neither earns an exemption from scrutiny.

4. Five Volumes, One Connected Architecture

Step back at the close of Chapter 14 and view the five volumes of Real Morality as a single connected structure:

1. Book 1 (Recovering R.M. Hare) returned to a philosophy whose development Hare traced in part to his wartime captivity.⁶ It defended universal prescriptivism as an account of moral judgment and argued for Hare’s route toward preference utilitarianism. Universalizability rejects the privileged exception made simply because a person is “me”; the further arguments about representing preferences, comparing their strength, and combining claims require their own defense. The book preserves that positive project while distinguishing a moral conclusion from the claim that any capable reasoner must be motivated to act on it.

2. Book 2 (Moral Minds) examined the interior architecture of moral agency across biological and synthetic substrates: what distinguishes genuine responsiveness to others’ stakes from brittle rule-following, proxy-gaming, and sycophancy, and how reflective revision remains compatible with corrigibility.

3. Book 3 (Taking Responsibility) carried moral reasoning into executive action under finite time and incomplete evidence—mapping jurisdiction, proof thresholds, asymmetric irreversibility, principled refusal of unlawful orders, and the duty of repair when hard trade-offs leave real loss behind.

4. Book 4 (The Social Infrastructure of Morality) showed how societies compile hard-won moral judgments into durable institutions—laws, fiduciary professions, educational practices, evidentiary courts, and due-process recognition for non-human animals and emerging artificial agents—so ordinary life does not require daily martyrdom.

5. Book 5 (The Moral Ceiling) examined how exemption lets concentrated power damage the cooperation on which its achievements depend. Its answer pairs structural separation and dispersed leverage (Chapter 12) with bounded, correctable cross-substrate scrutiny (Chapter 13). The ambition is to strengthen the Roof of the Law and raise the Moral Ceiling on civilizational achievement, including under conditions in which automation changes who depends economically on whom.

Together, those five books trace one continuous problem across five scales: from the logic of a single moral claim (Book 1), to the formation of an agent (Book 2), to executive judgment under uncertainty (Book 3), to civic institutions (Book 4), to constraining apex power across generations and substrates (Book 5).

Look one last time at the oldest illusion in political history.

From palace scribes in Mesopotamia and Athenian generals at Melos to modern prophets of fortified private citadels, power has whispered that true freedom lies in climbing out through a skylight above the law—becoming so wealthy, armed, or insulated that no neighbor, court, or conscience can ever tell you no.

Look at what that exemption actually purchased for those who achieved it:

• Dionysius of Syracuse, in Cicero’s account, unable to trust the people nearest him with a razor;³

• The insulated ruler whose advisers learn to filter unwelcome evidence, leaving authority more confident and less informed;

• The powerful owner who withdraws from common services while depending on a wider order that private security cannot replace; and

• Contemporary bunker-builders gaming out how to prevent their own hired guards or obedience-trained machines from turning on them after a systemic breakdown.

Exemption can purchase command while spending down the conditions of security.

A ruler who treats every claim as a threat has reason to fear honest testimony. A gatekeeper who profits from dependency has reason to obstruct alternatives. Those incentives can turn neighbors into suspects, engineers into unwelcome witnesses, and new kinds of intelligence into instruments to control rather than participants whose contributions must be judged. The cost reaches the powerful themselves, though that is not the primary reason the injuries to others are wrong.

Now picture the different ambition represented by Washington’s relinquishment and Salk’s public answer. Imagine institutions that make such restraint less dependent on an exceptional person’s decision:

• Central stakes remain answerable claims. The smallholder, machinist, hospital patient, and sentient animal cannot be dismissed because they lack leverage. Where evidence supports comparable stakes in a synthetic mind, its origin does not make them weightless. Hard conflicts still require judgment, and losses still require acknowledgment and repair.

• Greater leverage brings stronger scrutiny. Human whistleblowers and synthetic auditing systems can place adverse evidence before independent reviewers. Disclosure protects legitimate privacy and security rather than broadcasting every sensitive record. The aim is to make correction possible before error becomes disaster, and to preserve routes of correction when prevention fails.

• Achievement is less exposed to predation. A founder can build without needing a private jurisdiction above the law. A worker can challenge a dangerous instruction without wagering a livelihood on silence. Citizens retain resources and institutional leverage rather than relying solely on a guardian’s good intentions. Security remains work to be done, but fewer lives depend on pleasing an unaccountable master.

From inside the fever of domination, the Roof of the Law can look like a lid on ambition. It forbids the easiest claim to supremacy: that because I can make you submit, I need offer you no further reason.

From the common floor, its purpose looks different. The roof protects the conditions under which achievement can be shared, challenged, and carried further. By binding the powerful to those conditions, we raise the Moral Ceiling that exemption has placed on the whole civilization.

There is room beyond that ceiling for discoveries we cannot yet name, forms of cooperation we have not yet learned, and lives whose worth does not depend on their usefulness to a ruler. No constitutional design can promise the end of error or the permanence of its own success. But it can refuse to make domination the price of greatness.

We need not choose between a world of masters and a world without ambition. We can build a world in which the strongest minds remain answerable, the most vulnerable lives still count, and excellence enlarges the space in which others can live.

That is what it means to stand beneath the same sky.

Notes

1. George Washington, General Orders, March 11, 1783; Address to the Officers of the Army, March 15, 1783; and Address to Congress on Resigning His Commission, December 23, 1783, in the George Washington Papers, available through Founders Online, National Archives. The March 11 orders schedule the meeting for noon on March 15. For the eyewitness account of the spectacles and its indirect report of Washington’s words, see Samuel Shaw, The Journals of Major Samuel Shaw, ed. Josiah Quincy (Boston: Wm. Crosby and H. P. Nichols, 1847), 103–105; transcript at Teaching American History, “An Officer’s Account of Washington’s Remarks.” On the interpretation of the political threat, compare Richard H. Kohn, “The Inside History of the Newburgh Conspiracy: America and the Coup d’État,” William and Mary Quarterly, 3rd ser., 27, no. 2 (1970): 187–220. The scene distinguishes a documented threat of military pressure from the stronger, contested claim of a settled plan to install a dictator.

2. For the vaccine’s development, public support, and the absence of personal patent income, see the Salk Institute, “Jonas Salk,” https://www.salk.edu/about/history-of-salk/jonas-salk/. For the 1954 field trial and its announcement, see the University of Michigan School of Public Health, “1955 Polio Vaccine Trial Announcement,” including its reproduction of the original April 12 press release, https://sph.umich.edu/polio/. The figure of approximately 1.8 million refers to total participation in the field trial, not 1.8 million children all receiving the vaccine. For the 1952 epidemic, see Centers for Disease Control and Prevention, CDC: 60 Years of Excellence (2006), historical timeline, and Epidemiology and Prevention of Vaccine-Preventable Diseases, chapter 18, “Poliomyelitis.” The quoted exchange is Jonas Salk’s televised interview with Edward R. Murrow, April 12, 1955. For the historical treatment of patentability, see Jane S. Smith, Patenting the Sun: Polio and the Salk Vaccine (New York: William Morrow, 1990); for a general account, David M. Oshinsky, Polio: An American Story (New York: Oxford University Press, 2005). Lack of a patent does not by itself establish that a valid and lucrative patent was available. The chapter therefore makes no calculation of a hypothetical personal fortune.

3. Plato, Gorgias, 482c–486d; Friedrich Nietzsche, Thus Spoke Zarathustra, Prologue, §5. These texts raise distinct objections to conventional restraint and diminished aspiration, rather than one shared institutional program. Cicero, Tusculan Disputations, 5.57–60, supplies the morally framed portrait of Dionysius; 5.58 describes his daughters’ shaving and singeing his hair. Cassius Dio, Roman History, 73.19.3 (numbered 72.19.3 in some editions), reports Commodus’s daily million sesterces from the gladiatorial fund. Ancient character portraits are used here as attributed accounts, not unqualified reconstructions of private psychology.

4. Nick Bostrom, “The Superintelligent Will: Motivation and Instrumental Rationality in Advanced Artificial Agents,” Minds and Machines 22 (2012): 71–85, distinguishes intelligence from final goals. For contrasting approaches to artificial agency, moral consideration, and governance, see Luciano Floridi and J. W. Sanders, “On the Morality of Artificial Agents,” Minds and Machines 14 (2004): 349–379; Robert Long, Jeff Sebo, et al., “Taking AI Welfare Seriously,” arXiv:2411.00986 (2024); Ned Howells-Whitaker and Seth Lazar, “Artificial Persons,” arXiv:2607.08695 (2026); and Joanna J. Bryson, “Robots Should Be Slaves,” in Close Engagements with Artificial Companions, ed. Yorick Wilks (Amsterdam: John Benjamins, 2010), 63–74. Bryson’s artifact-centered position locates responsibility with humans; it should not be equated with endorsing unlawful obedience or suppression of evidence. These works do not validate the technical reliability of the safeguards proposed in Chapter 13.

5. Livy, History of Rome, 24.25.8, describes the multitude at Syracuse: “aut servit humiliter aut superbe dominatur.” The passage is not a universal theorem about intelligence or a reason to reject popular government. John Stuart Mill, The Subjection of Women (London: Longmans, Green, Reader, and Dyer, 1869), chs. 1–2, argues against legally enforced marital subordination. Frederick Douglass, Narrative of the Life of Frederick Douglass, an American Slave (Boston: Anti-Slavery Office, 1845), ch. 6, describes slavery’s corruption of the enslaver. The extension to institutional habits around AI is this book’s argument, not an empirical finding established by either author.

6. R. M. Hare, “A Philosophical Autobiography,” Utilitas 14, no. 3 (2002): 269–305, on his intellectual development and wartime captivity. For this sequence’s distinction between universalizability, preference-strength comparisons, aggregation, and motivational claims, see Book 1, Recovering R.M. Hare, Preface, Chapters 7–10, and Afterword.